Every business that has ever run a planning day knows the pattern. Two days offsite, a whiteboard covered in initiatives, genuine energy in the room, a document circulated the following Monday. Six weeks later nobody can name the third priority.
The plan was not wrong. The strategy was not badly reasoned. What failed was everything between the document and the daily work.
This gap between what gets decided and what gets done is the most consistently documented problem in management research, and it is worse in small companies than in large ones, for reasons that have nothing to do with capability.
The size of the problem
The numbers are grim and remarkably consistent across sources.
Harvard Business Review research puts the proportion of well-formulated strategies that fail at execution at 67%. Kaplan and Norton's work found that up to 90% of strategic plans are not executed successfully. McKinsey's research on transformations found 70% fail at the execution stage rather than the design stage.
Perhaps the most telling figure: only around 10% of C-level executives report implementing two-thirds or more of their core strategic initiatives in any given year. Executives themselves estimate they lose close to 40% of a strategy's potential value to execution breakdown.
These figures come from studies of large organisations, and the temptation is to assume small businesses do better because they are closer to the work. In practice the opposite is often true. Large organisations at least have planning functions, quarterly review structures and someone whose job it is to notice when an initiative has stalled. A twelve-person company has none of that. It has a founder who remembers the plan, and a team that heard it once.
Why plans stall
The reasons are unglamorous and almost entirely structural.
Too many priorities is the first and largest. The research on OKRs is unambiguous here: teams running one or two company objectives per quarter are roughly twice as likely to achieve them as teams running three or more. This is not a statement about ambition. It is a statement about attention. Every additional priority does not divide focus proportionally, it divides it disproportionately, because each one carries coordination cost as well as execution cost.
Most planning sessions produce eight to twelve initiatives because that feels like an appropriate output for two days of work. The quantity is the problem.
Disconnection between company goals and team work is the second. Research on OKR implementation found 65% of teams admitting their objectives are not directly linked to company goals. Read that again. Two thirds of teams running a goal framework are running goals that do not connect to the thing the goals exist to serve.
This happens because cascading is treated as a formatting exercise rather than a reasoning one. A team takes the company objective, writes something adjacent to it, and both documents exist without either constraining the other.
Ambiguous ownership is the third. An initiative owned by "the leadership team" or "sales and marketing" is owned by nobody. The rule that survives contact with reality is one owner, one due date, one outcome. Shared ownership sounds collaborative and functions as diffusion of responsibility.
No review rhythm is the fourth and most fixable. A plan reviewed quarterly is a plan that can drift for eleven weeks before anyone notices. By the time the quarterly review happens, the honest options are to explain the miss or to quietly reframe the goal. Neither produces execution.
Measurement that arrives too late is the fifth. If the only signal that an initiative is failing is the outcome metric, you learn about failure after it has finished happening. Revenue is a lagging indicator. So is customer retention. So is almost every metric that appears on a strategic plan.
The operating rhythm that closes the gap
The fix is not a better plan. It is a cadence that forces the plan to meet reality on a schedule.
The structure that works has four frequencies, each making a different class of decision. The distinction matters, because most struggling businesses have meetings at all four frequencies that all do the same thing: report status.
The weekly review is where execution actually lives. Thirty to forty-five minutes, same day, same time, same agenda. It opens with the actions committed last week and whether they happened, not with a discussion of what is happening generally. That ordering is the single most important design choice in the whole system. A weekly meeting that starts with accountability produces execution. A weekly meeting that starts with discovery produces conversation.
The agenda that holds up covers four questions. What changed in the numbers since last week. What is blocked. What decision is needed today. Who owns what before the next review. Anything that does not fit those four questions belongs in a different meeting.
The monthly review looks at whether the quarter is on track and whether the initiatives chosen are still the right ones. This is where you kill things. Small businesses are generally good at starting initiatives and terrible at stopping them, and the monthly review is the designated place for that decision.
The quarterly review sets the next set of priorities and honestly assesses the last set. The honesty part requires structure, because the natural human tendency is to grade generously. Scoring key results numerically before discussing them helps, because it forces the assessment before the narrative.
The annual review handles direction, positioning and the questions that are too large for a quarter.
The critical discipline across all four is that every meeting produces a decision. If a meeting ends without one, it should not have been held. Status can be written down and read. Decisions need people in a room.
Making goals that can actually be tracked
A goal you cannot check weekly is a goal that will drift for a quarter.
The practical test is whether you can tell, on any given Wednesday, if this goal is on track. "Improve customer satisfaction" fails that test. "Reduce first response time below four hours" passes it, because you can look.
This is why leading indicators matter more than the outcome metrics that populate most strategic plans. Revenue tells you what happened. Qualified meetings booked tells you what is about to happen, and you can act on it while acting is still possible.
For each objective, the useful pairing is one outcome metric that defines success and two or three leading indicators you can watch weekly. The outcome metric goes on the board. The leading indicators go in the weekly review.
The other discipline worth adopting is writing the target and the current value at the same time. A goal stated without a baseline cannot be assessed for ambition, and a surprising proportion of goals set in planning sessions turn out on inspection to be either already achieved or physically impossible.
The first cycle is meant to be bad
One finding from OKR research is genuinely encouraging and rarely mentioned. Teams in their first or second cycle average around 51% completion. By the fifth cycle and beyond, the same organisations average 79%.
The skill being developed is not goal setting. It is calibration. Knowing what your team can actually deliver in a quarter is learned experimentally, and there is no way to shortcut it. Every organisation sets absurd goals in its first cycle.
This matters because most goal frameworks are abandoned after one or two cycles, on the reasoning that they did not work. The data suggests they were working exactly as expected and were dropped one cycle before the benefit arrived.
It also matters for how you respond to a bad first quarter. The correct response is to keep the system and adjust the ambition. The common response is to conclude the system is bureaucratic and return to the previous approach, which was not being measured at all and therefore never appeared to fail.
Related to this: more than half of companies using structured goal frameworks have been using them for under three years, and around 87% report the approach met or exceeded expectations. The failure rate is not in the method. It is in the abandonment.
What small companies should do differently
Most goal-setting advice is written for organisations with a few hundred people, and the translation to twelve people is not obvious.
Run one company objective, not three. At small scale the coordination benefit of multiple parallel priorities does not exist, because the same people are working on all of them. Three objectives in a twelve-person company is one objective with three names.
Skip the cascade entirely below about thirty people. Team-level objectives that ladder to company objectives make sense when teams are genuinely distinct units of work. Below that size, individual owners against company key results works better and creates far less documentation.
Put the review in an existing meeting. The failure mode for small teams is adding a new ceremony, which competes with actual work and gets cancelled the first busy week. Attach the weekly review to a meeting that already happens and already has attendance.
Keep it visible without a project. If the goals live in a document somebody has to open, they will be seen at planning and at review and never in between. On a wall, in a shared channel pinned to the top, or in a tool the team already opens daily. The specific medium matters less than the fact that seeing them requires no decision.
Write down what you are not doing. This is the most useful and least practised part of small-company planning. A short list of things explicitly deferred this quarter prevents the slow accumulation of side projects that is the primary way small companies lose focus.
Running a planning session that produces less
Given that too many priorities is the primary cause of execution failure, the planning session needs to be designed to reduce rather than to generate. Most are designed to do the opposite.
The standard format opens with brainstorming, which produces volume, and then attempts to prioritise a long list under time pressure at the end of day two when everyone is tired and reluctant to argue against a colleague's idea. The output is a compromise list containing everything anyone cared about.
A better structure inverts it. Start with the previous period, honestly assessed and scored before it is discussed. Then spend the bulk of the time on a single question: what is the most important thing that has to be true by the end of this quarter. Not the list of things worth doing, which is always long, but the one outcome that would make the quarter a success.
Argue about that one thing for as long as it takes. It is genuinely difficult and the difficulty is the point. A leadership team that cannot agree on a single most important outcome does not have a prioritisation problem, it has a strategy disagreement that has been hidden by the practice of listing everything.
Once the primary outcome is settled, work backwards to the two or three measures that would show progress, and then to the small number of initiatives that move those measures. This ordering means initiatives are derived from the outcome rather than collected and then justified.
Finish by writing the not-doing list. Take everything raised during the session that did not make the cut and record it explicitly as deferred. This does two things. It stops good ideas from being lost, which is the usual objection to narrowing. And it creates a document you can point at in week six when someone starts working on one of them.
Half a day is usually enough. Two-day offsites generate output proportional to their length, which given the diagnosis above is a problem rather than a benefit.
Diagnosing your own execution gap
Four questions, answered honestly, will locate the problem.
Can every person in the business name the top priority this quarter without checking? If not, the problem is communication frequency, not the plan.
Does every active initiative have exactly one named owner and a date? If not, the problem is ownership structure.
When did you last stop an initiative that was not working? If the answer is never, the problem is that your review meetings are reporting rather than deciding.
Do you find out an initiative has stalled within a week, or within a quarter? If it is a quarter, the problem is measurement latency and review cadence.
Most businesses find their answer in the last two. Starting things is easy and social. Stopping them requires someone to say the initiative is not working, in a forum designed for that conversation, at a point when stopping still saves something.
When the goal was wrong
A situation the frameworks handle poorly: halfway through the quarter it becomes clear the goal itself was misconceived. Not difficult, not behind, but aimed at the wrong thing because the market moved or an assumption turned out to be false.
The orthodox advice is to hold the goal to the end of the period for the sake of discipline. This is wrong, and following it destroys confidence in the whole system, because everyone can see the team working towards something that no longer matters.
The workable rule is that goals can change, but changing one is a decision that gets made explicitly, in the monthly review, with the reason recorded. What corrodes trust is not change, it is silent change: the goal quietly rewritten so that the result looks better, or dropped without acknowledgement.
Record it as changed and record why. At the end of the year you will have a list of the goals you altered mid-flight, and the pattern in that list is usually the most valuable planning information you have. If most changes trace to the same source, such as consistently underestimating delivery timelines or repeatedly betting on a channel that has not worked, that is the thing to fix next year.
The same logic applies to goals that turn out to be far too easy. A key result hit in week three was not a goal, it was a forecast. Note it, replace it, and calibrate upwards next quarter.
The part that is genuinely hard
Everything above is mechanical, and mechanical problems have mechanical fixes. The hard part is different.
Execution discipline requires someone to hold the rhythm when it is inconvenient, and it is always inconvenient. The weekly review will feel unnecessary in a busy week, which is precisely the week it does the most work. The initiative that should be killed will have a sponsor who believes in it. The goal that was missed will have a reasonable explanation.
The businesses that close the execution gap are not the ones with better frameworks. They are the ones where the review happens in the busy week, the initiative gets killed despite the sponsor, and the missed goal is recorded as missed before the explanation is discussed.
That is a cultural property, not a software one. What software can do is remove the friction that gives people an excuse: making the numbers visible without someone compiling them, making ownership explicit without someone chasing it, and making the review agenda assemble itself.
Empiraa GPS is built for exactly that layer, keeping goals, KPIs, actions and the meeting rhythm in one place so the weekly review starts with what happened rather than with someone finding out.
Frequently asked questions
What is the strategy execution gap?
The strategy execution gap is the difference between what an organisation decides to do and what it actually does. Harvard Business Review research puts the proportion of well-formulated strategies that fail at execution at 67%, while work by Kaplan and Norton found up to 90% of strategic plans are not executed successfully. The failure is generally located in implementation rather than in the quality of the strategic thinking.
How many company goals should a small business set per quarter?
One or two. Research on goal frameworks indicates teams running one or two company objectives per quarter are roughly twice as likely to achieve them as teams running three or more. In businesses under about thirty people the same individuals work across all objectives, so additional priorities divide attention rather than distributing work.
How often should goals be reviewed?
Weekly for execution and monthly for direction. A weekly review of thirty to forty-five minutes covering what changed, what is blocked, what decision is needed and who owns what before next week is where execution actually happens. Quarterly review alone allows an initiative to drift for eleven weeks before anyone notices it has stalled.
Why do OKRs fail in small companies?
The most common causes are too many objectives, objectives that are not connected to the work teams actually do, and abandonment after one or two cycles. Research indicates 65% of teams admit their objectives are not directly linked to company goals. On abandonment, teams in their first or second cycle average around 51% completion while the same organisations reach 79% by the fifth cycle, so frameworks are often dropped shortly before the benefit arrives.
Should a goal be changed partway through a quarter?
Yes, when the goal has become misconceived rather than merely difficult. The important condition is that the change is made explicitly in a scheduled review with the reason recorded, rather than quietly rewritten. Silent revision is what damages confidence in a goal system, not revision itself.
What is the difference between a leading and a lagging indicator?
A lagging indicator such as revenue or retention reports what has already happened. A leading indicator such as qualified meetings booked or proposals issued moves earlier and can still be acted on. Most strategic plans are populated entirely with lagging indicators, which is why problems are discovered too late to address within the period.
The smallest useful change
If the whole system feels like too much, start with one thing.
Pick the single most important outcome for this quarter. Name one owner. Identify two leading indicators you can check on a Wednesday. Put fifteen minutes in the calendar every week to look at them and decide one thing.
That is not a strategy execution system. It is one loop. But it is the loop that everything else is built from, and running it for a quarter will teach you more about your execution gap than another planning day will.
